【摘要】 准确可靠的电动车辆续驶里程估计结果能够为驾驶员安排出行提供依据,提高驾驶员对于电动车辆的驾驶信心。目前剩余里程估计方法往往忽略了交通环境复杂多变的特性,这使得估计结果与车辆实际可达范围存在较大出入。该文提出了基于动态交通信息的电动车辆剩余里程估算方法,利用动态随机路网模型反映交通环境的动态性与随机性,以保证估计结果能够充分考虑到交通环境随时间的变化波动。通过基于随机模拟的遗传算法对电动车辆行驶过程的能量消耗进行估计,并利用随机规划模型分析电动车辆的可达范围。研究结果表明:该期望值模型的估计结果能够有效控制高估与低估比率的幅值,同时相关机会模型的估计结果能够最大限度地抑制高估比例,得到保守的可达范围,因此可最终得到"内圈"与"外圈"的剩余里程估计结果。该剩余里程估计方法充分考虑了交通环境的动态性与随机性,相比于传统的里程估计方法,估计结果更为合理,且估计结果波动降低。更多还原

【Abstract】 Accurate and reliable remaining mileage estimates for electric vehicles(EV)will assist the driver in planning trips and enhance the driver’s driving confidence. Traditional remaining mileage estimation methods have overlooked the complex changes due to traffic conditions,which can lead to large differences between the estimates and the real reachable area.A remaining mileage estimation method based on dynamic traffic information is developed in this paper.The model depicts the complex features of the traffic system with a stochastic,time-dependent traffic network model that improves estimation accuracy by taking the variable traffic conditions into consideration.The EV energy consumption is predicted by a stochastic simulation based genetic algorithm,with the accessible area evaluated by a stochastic programming method.Tests show that the accessible area estimates of this expectation model effectively reduces overestimates and underestimates, while the related chance model controls the overestimates to give conservative estimates.These are combined in an "inner-outer"layer estimate.The remaining mileage estimation method described here takes dynamic and stochastic traffic conditions into consideration to give more realistic estimates with smaller variations than the traditional mileage estimation method.更多还原